Giving patients hard numbers about the likelihood of experiencing side effects from a new medication makes them more likely to take the drug, according to a new study fromthe UNC Eshelman School of Pharmacy. Conversely, giving them similar information about the likelihood of benefiting from a drug has the opposite effect, making patients less willing to take it.
A team led by Susan Blalock, Ph.D., M.P.H., a professor in and vice chair of the Division of Pharmaceutical Outcomes and Policy, asked nearly one thousand people to review information about the risks and benefits associated with a hypothetical drug. Some participants were given information that only listed the purpose of the drug and possible side effects. Others also received information about the number of people who were likely to benefit from the medicine or experience side effects because of it. They were then asked whether they would take the medication and to evaluate its effectiveness.
An example of non-numeric information is a statement that says taking a drug helps prevent heart attacks and stroke and that possible side effects include dry mouth, headache or muscle damage. This is similar to what is provided to patients today as mandated by the Food and Drug Administration. In contrast, adding numerical information might tell patients that if 100,000 people took the medicine, 2,250 fewer would have a heart attack or stroke, or that, out of 100,000 patients, 14,000 experienced dry mouth, 7,000 had headaches and two suffered muscle damage.
“Our findings suggest that non-numeric side-effect information conveys the gist that the medication can cause harm, reducing patients’ willingness to use the medication. Non-numeric benefit information has the opposite effect,” Blalock said. “Presenting side-effect and benefit information in non-numeric format appears to bias decision-making in opposite directions. Providing numeric information for both benefits and side-effects may enhance decision-making.”
Numeric risk information increased the study participants’ willingness to use a hypothetical medication. Numeric benefit information decreased their willingness to use the medication. The findings were published in the journal Patient Education and Counseling.
“Side-effect information presented in a non-numeric format conveys the gist that medication can cause harm, which in turn decreases willingness to take it,” Blalock said. “Presenting benefit information the same way, however, has the opposite effect. Providing numeric medication risk and benefit information appears to enhance decision-making.”
Raising an Ethical Question
Currently most written medical information contains little numeric information about side effects and benefits. Findings from this study suggest that presenting information on side effects in numeric format and keeping benefit information in non-numeric format will likely lead to considerable bias in favor of using medication. Presenting benefit information in numeric format as well, however, will likely increase patient unwillingness to initiate and continue therapy, which in turn could generate ethical dilemmas for health-care providers.
“Providing numeric benefit information may decrease patients’ willingness to try a medication or take it correctly,” Blalock said. “This puts a health-care provider is a position where withholding information from a patient could make the patient more willing to proceed with a certain course of treatment.”
The Study
In the study, the international research team presented the 999 participants with information about side effects in one of three formats and information about benefits in one of five formats. For both, one of the formats contained only non-numeric information, while the rest contained numeric information.
The researchers recruited participants through Amazon Mechanical Turk, a web service that provides a human workforce to do tasks that computers are not yet able to do, with the goal of using fuzzy trace theory to study how displaying medication side effects and benefits numerically and non-numerically affected willingness to take a hypothetical medication.
Fuzzy trace theory explains how people make decisions that involve uncertainty. It suggests that when an individual is exposed to a meaningful stimulus such as written medical information, two representations are encoded in his or her memory: verbatim representations, which capture the exact words, numbers or images of a stimulus, and gist representations, which capture its essential meaning, including the emotions associated with the meaning.
Authors
- Susan J. Blalock, Ph.D., M.P.H., a professor in the Division of Pharmaceutical Outcomes and Policy at the UNC Eshelman School of Pharmacy;
- Adam Sage, a graduate student in the Division of Pharmaceutical Outcomes and Policy advised by Delesha Carpenter, Ph.D., M.S.P.H., at the UNC Eshelman School of Pharmacy;
- Michael Bitonti, a Pharm.D. candidate at the UNC Eshelman School of Pharmacy;
- Payal Patel, a Pharm.D. candidate at the UNC Eshelman School of Pharmacy;
- Rebecca Dickinson, Ph.D., a lecturer in adult nursing in the School of Healthcare, University of Leeds; and
- Peter Knapp, Ph.D., a senior lecturer in the Department of Health Sciences and the Hull York Medical School, University of York.